Commit History

fix for accelerate env var for auto bf16, add new base image and expand torch_cuda_arch_list support (#1413)
da265dd
unverified

winglian commited on

deprecate: pytorch 2.0.1 image (#1315) [skip ci]
a359579
unverified

Nanobit commited on

run the docker image builds and push on gh action gpu runners (#1218)
aaf54dc
unverified

winglian commited on

drop py39 docker images, add py311, upgrade pytorch to 2.1.2 (#1205)
74c72ca
unverified

winglian commited on

support for cuda 12.1 (#989)
37820f6
unverified

winglian commited on

Dockerfile torch fix (#987)
161bcb6
unverified

winglian commited on

add docker images for pytorch 2.10 (#697)
7f2618b
unverified

winglian commited on

Prune cuda117 (#327)
2c37bf6
unverified

winglian commited on

don't use the gha cache w docker
c5df969

winglian commited on

use pytorch 2.0.1
c58034d

winglian commited on

11.7.0 nvidia/cuda docker images are deprecated, move to 11.7.1
a10da1c

winglian commited on

don't fail fast
d35278a

winglian commited on

Fix base builder, missing tags
e0011fd

winglian commited on

define python version (3.10) explicitly as string in yaml
c5b0af1

winglian commited on

py310, fix cuda arg in deepspeed
c43c5c8

winglian commited on

cleanup from pr feedback
48612f8

winglian commited on

default to qlora support, make gptq specific image
6ef96f5

winglian commited on

move CUDA_VERSION_BNB arg inside of stage build scope
e43bcc6

winglian commited on

bnb fixes
21f17cc

winglian commited on

update docker to compile latest bnb to properly support qlora
312b8d5

winglian commited on

cuda, pytorch matrix for base builds
f950a88

winglian commited on

docker layer caching, build w axolotl from base build
990bec6

winglian commited on

build base separately
2734e3f

winglian commited on